Big Data Isn’t a Thing; Big Data is a State of Mind
“Big Data is dead.” “Big Data is passé.”
“We no longer need Big Data; we need Machine Learning now.”
As we end 2017 and look forward to big (data) things in 2018, the most important lessons of 2017 – in fact, maybe the most important lesson going forward – is that Big Data is NOT a thing. Big Data isn’t about the volume, variety or velocity of data any more than car racing is about the gasoline. Big Data is a state of mind. Big Data is about becoming more effective at leveraging data and analytics to power your business models (see Figure 1).
Big Data is a State of Mind
Big Data is about improving an organization’s ability to leverage data and analytics to power their business models; to optimize key business and operational use cases; reduce security and compliance risk; to uncover new revenue opportunities; and create more compelling, differentiated customer engagements. The technical components – building blocks – of a “big data state of mind” include:
- Data: Ability to collect and aggregate detailed data from a wide variety of data sources including structured (tables, relational databases), semi-structured (logs files, XML, JSON) and unstructured data sources (text, video, audio, images).
- Analytics: Ability to leverage advanced analytics (data science, deep learning, machine learning, artificial intelligence) to uncover customer, product, service, operational, and market insights.
These are important technology building blocks, but by themselves, they provide NO business or financial value. These are necessary but not sufficient capabilities for driving the most important aspect of Big Data – Data Monetization!
Big Data is About Data Monetization
Big Data is about exploiting the unique characteristics of data and analytics as digital assets to create new sources of economic value for the organization. Most assets exhibit a one-to-one transactional relationship. For example, the quantifiable value of a dollar as an asset is finite – it can only be used to buy one item or service at a time. Same with human assets, as a person can only do one job at a time. But measuring the value of data as an asset is not constrained by those transactional limitations. In fact, data is an unusual asset as it exhibits an Economic Multiplier Effect, whereby it never depletes or wears out and can be used simultaneously across multiple use cases at near zero margin cost. This makes data a powerful asset in which to invest (see Figure 2).
Understanding the economic characteristics of data and analytics as digital assets is the first step in monetizing your data via predictive, prescriptive and preventative analytics.
See the blog series at “Determining Economic Predicted Value of Data (EPvD) Series” for more insights about how organizations can exploit the unique economic characteristics of data and analytics as digital assets.
Big Data is a Business Discipline
Leading organizations that embrace digital transformation see data and analytics as a business discipline, not just another IT task. And tomorrow’s business leaders must become experts at leveraging data and analytics to power their business models. The most valuable companies today (from a market cap perspective) are those organizations that are mastering the use of Big Data (with artificial intelligence, machine learning, deep learning) to derive and drive new sources of value (see Figure 3).
At the University of San Francisco, I teach the “Big Data MBA” where I am educating tomorrow’s business leaders how to embrace data and analytics as the next modern business discipline. A Master of Business Administration (MBA) provides theoretical and practical training to teach business leaders important business disciplines such as accounting, finance, operations management and marketing. We want to treat analytics as a similar business discipline.
Data Science is the Data Monetization Engine
Data Science is used to identify the variables and metrics that might be better predictors of business and operational performance, and to quantify cause-and-effect in order to predict likely actions and outcomes; prescribe corrective actions or recommendations; prevent costly outcomes; and continuously learn and adapt as the environment changes.
To do that, data scientists need to learn a wide variety of statistical, data mining, deep learning, machine learning, and artificial intelligence techniques and tools (see Figure 4).
Data monetization requires close collaboration with business stakeholders who own the important responsibility of setting the business and analytics strategy. These stakeholders also unambiguously define the hypotheses to be tested, and articulate how the resulting analytic outcomes will be operationalized and monetized. The key to enlisting business leadership is to turn them into “Citizens of Data Science” and to teach them to “Think Like a Data Scientist.”
- Use case identification, validation and prioritization that begins with an end in mind.
- Develop personas for each key business stakeholder and constituent to understand their responsibilities, key decisions, and impediments to success.
- Brainstorming variables and metrics that might be better predictors of performance.
- Creating actionable, prescriptive analytic insights and recommendations that drive measurably better operational and business decisions.
- Articulating how the analytic outcomes will be operationalize or put into action.
Check out the infographic “Think Like A Data Scientist” for more information. It also includes a workbook that guides the “thinking like a data scientist” process.
A Big Data State of Mind
One of my favorite articles (So, What Is Machine Learning Anyways?) does a great job of summarizing the important relationship between Big Data and Machine Learning:
- Big Data started when the Internet created a treasure trove of website and search data. Today that data has been augmented by social media, mobile, wearables, IOT, and even microphones and cameras that are constantly collecting information.
- With so much data readily available, machine learning provides a method to organize that data into meaningful patterns. Machine learning sorts through those troves of data to discern patterns and predict new ones.
- Machine learning plays a key role in the development of artificial intelligence. Artificial intelligence refers to a machine’s ability to perform intelligent tasks, whereas machine learning refers to the automated process by which machines weed out meaningful patterns in data. Without machine learning, artificial intelligence as wouldn’t be possible.
Though there are many critical building blocks associated with Big Data, the leading organizations are quickly realizing the Big Data isn’t a thing.
Big Data is a mindset about transforming business leadership to become more effective at leveraging data and analytics to power the organization’s business models (see Figure 5).
So, how effective is your organization at leveraging #BigData and #MachineLearning to power your business models and create an intelligent organization?